OpenAI’s New o1's Mensa IQ Test Result

https://twitter.com/maximlott

OpenAI’s new o1 model has been highlighted as a significant advancement in AI intelligence, particularly after being tested with the Norway Mensa IQ test.

According to Maxim Lott, who conducted the analysis, the o1 model outperformed other AI systems in this evaluation. Lott expressed surprise at this progress, especially given the lack of public advancements in AI over the previous six months.

He noted that the Mensa IQ test used for this assessment was designed specifically for offline use and is not included in any AI training data, which may contribute to lower scores compared to public IQ tests.

The full analysis can be found on his platform, maximumtruth.org

OpenAI’s o1 Model Achieves High IQ Score

• OpenAI’s new “o1” model scored 25 out of 35 questions correct on the Norway Mensa IQ test, suggesting an IQ around 120
• This represents a significant leap in AI reasoning ability compared to previous models like ChatGPT.
• The model was able to detect complex patterns and provide logical explanations for its answers on challenging IQ…